26 July 2000. Updated 11 October 2000
Interviewed by Chris Weihl
ARF: Dr. Mayeux, your research has revealed several risk factors
that are associated with AD. What do you think are the most important risk
factors for the development of AD?
Mayeux: I think the most significant risk factors are genetic.
The evidence accumulated over the last decade is pretty clearly pointing
to a genetic etiology for the disease. For example, there is no question
that in early onset familial AD, mutations in 3 key genes (APP, PS1, PS2)
almost invariably result in disease. What remains to be established is
whether there are other mutations, as yet unidentified, in other genes that
explain the remaining cases of early onset, familial AD (EOAD). In addition,
there are sporadic cases of EOAD that are unexplained and that don't appear
to have a strong family history. Because we have learned that presenilin1
may have new mutations, it is possible that something like this causes EOAD
in the remaining cases.
ARF: Are these presenilin mutations somatic or germ-line mutations?
Mayeux: Germ line.
ARF: What about late onset AD?
Mayeux: With late onset AD, the APOE story has been truly remarkable.
There is no debate that APOE-e4 is a strong risk factor for late onset
AD (familial and sporadic), but what we don't know is exactly how APOE-e4
increases the risk. We also don't know whether the polymorphism is modified
by other factors—either genes or environmental—that also contribute
to the risk of AD. There is weak evidence that head trauma or smoking may
interact with APOE in some curious way. There is currently no other known
polymorphism in any other gene that is associated with late onset AD to
the degree that APOE-e4 is associated.
The alpha-2-macroglobulin polymorphism remains controversial. There are
hints from other large studies suggesting that there are other loci in late
onset AD. However, no specific mutation or polymorphisms in genes have been
identified. It may just be a matter of time and luck before that happens.
Now, having said that, a number of case control studies have looked at
genes that might be biologically related to AD. The IL-1 polymorphism on
chromosome 2 has been the subject of three reports that now suggest an association
with AD. The same group of investigators has been involved with each study,
but they are in different populations. This is pretty interesting and it
may hold up. If it is true, it will be first polymorphism associated with
AD that was found using this non-traditional study design. That in and of
itself would be pretty remarkable. Again this points to a disease in which
genetic influences are perhaps the strongest risk factors and may account
for the highest contribution to disease risk.
What is intriguing about all the genes that have been implicated in AD
is that they all point toward a disturbance in amyloid metabolism from amyloid
precursor protein (APP) down to the deposition of A-beta. So that I think
is very simple and biologically sensible.
ARF: What about environmental factors such as head trauma?
Mayeux: The head trauma story is controversial and inconsistent.
About half the studies show that there is a strong association between AD
and head trauma, while half the studies do not. If true, the association
is probably weak and the association of head injury to AD risk must be very
small. Head trauma may interact with genes such as APOE. In one study
we found an interaction between head trauma and APOE-e4 . However, others
could not confirm this study. So it is not yet clear whether environmental
factors, like head trauma, trigger AD among people who are genetically susceptible
or whether the genetic susceptibility is the driving the relationship.
ARF: Will treatments be aimed at these aberrant genes?
Mayeux: It look like scientifically based treatments are starting
to emerge. Obviously, if the primary cause of AD is the overproduction
of A-beta 42, as in EOAD, and if late onset AD is a result of faulty clearance
of A-beta 42, then targeting A-beta 42 would be the first logical step.
The amyloid vaccine being developed by certainly looks promising and could
directly attack A-beta 42. The animal studies looked very encouraging.
Vaccine use in people that are genetically susceptible will be an important
However, there is also encouraging data concerning PS1 as either "the"
gamma secretase or the active site for this enzyme. If gamma secretase
inhibitors currently in development prove to be safe and effective in diminishing
the amount of A-beta 42 generated, such a drug in people at high risk would
also seem logical. I imagine that we will eventually use genetic factors
to identify people who are at extremely high risk and then we will target
prevention therapies, either vaccines or drugs, which reduce the risk of
AD to a minimum. The first prototype will probably be drugs that target
the amyloid pathway. At least that is my guess.
ARF: What about current treatments? Will they be obsolete? For
example will cholinesterase inhibitors not be used in the future?
Mayeux: I don't know. It may be that once you get rid of the
amyloid or halt the amyloid deposition process, you still may have memory
impairment and we may need cholinesterase inhibitors as well. If I had
to make a guess, I would say that you see the use of cholinesterase inhibitors
decrease once preventive strategies are developed.
ARF: Is there still a need to map out more susceptibility genes?
Mayeux: Absolutely! The problem is that those contributing most
to AD have probably already been found. Probably the remaining genes that
the AD gene hunters are looking for are genes in which mutations or polymorphisms
are rare. These mutations and polymorphisms may account for only a small
proportion of disease. It is also possible that new allelic polymorphisms
modify existing AD genes by some, as yet unknown, gene-gene interaction.
In other words, I think it is going to be difficult.
On the other hand, seeking them out is very important since there still
are a very large number of people whom we cannot explain why they have AD.
The gene hunters and the epidemiologists interested in AD are beginning
a dialog. We should help in this effort. It is going to be important to
keep this line of research open.
As the biology of this disease continues to be worked out, more targets
along the amyloid cascade may become available. Perhaps tau or alpha synuclein
will be important targets too. It is conceivable that we will create genetic
profiles to assess risk in people. It won't be a single gene, but a cluster
of genes that we will have to understand. And then probably some of these
inconsistencies in environmental factors, such as smoking and head injury
will start to become clearer. If I had to make a prediction, I would say
that genes will tell the story and environmental factors will be modifiers
of genetic risk but only subtle modifiers and not major modifiers.
ARF: Could you please describe the goals and aims of your own
Mayeux: I'm trained both in neurology and epidemiology, so most
of my research is centered on the epidemiology of AD. We were the first
to show in a prospective study that estrogen in post-menopausal women might
be useful in disease protection . We have also looked at education and
head injury all in prospective studies . These are at an advantage
to cross-sectional or case control studies.
Our strategy is to identify healthy populations and test hypotheses such
as, what is the effect of estrogen use among post-menopausal women on AD
risk? What is the effect of having an APOE-e4 allele in a healthy person
over a 5-7 year period? Primarily large scale prospective longitudinal
epidemiological studies examining both genetic and environmental risk factors
for AD. Paying attention to risk factors, either genetic or environmental,
that increase risk. Also examining behaviors, drugs or dietary supplements
that lower risk. We have also worked in various ethnic groups to determine
how APOE-e4 and other genes segregate in those groups and how they affect
ARF: Has the data regarding estrogen use been demonstrated in
Mayeux: There are about 5 observation studies now showing basically
the same results that post-menopausal women who use estrogen have lower
risk of AD. Currently, one of my colleagues, Mary Sano, is leading a study
in which women who have a family history of AD are randomized to estrogen
or placebo to see whether or not we can actually reduce the risk of the
ARF: Is the diagnosis of AD currently accurate enough that those
studies can be performed or do we need to wait for post-mortem analysis.
Mayeux: There is myth that the only way you can diagnose AD is
through post-mortem histopathology. This may, in part, be correct but neuropathologists
do not always agree on the criteria to make this diagnosis at death. This
make the concordance between the clinical diagnosis and the pathological
diagnosis difficult. Nonetheless, a clinico-pathologic study from the 27
AD centers yielded an overall accuracy of ~90% at least at AD research centers
. The clinical diagnosis has high sensitivity, but lacks specificity. Physicians
seldom confuse dementia with normal cognitive function, but they can incorrectly
attribute dementia to AD when another diagnosis is present. So I would say
the accuracy of clinical diagnosis is pretty good. It has gotten better
with MRI, neuropsychological testing and supporting laboratory studies.
Someday soon we may also include genetic information.
ARF: What is most confounding diagnosis?
Mayeux: I think cerebrovascular disease. The dilemma is because
cerebrovascular disease and AD often coexist. Some investigators have postulated
that cerebrovascular risk factor increase the risk of AD, while others do
not support that view. There are data that suggest that if you have AD and
then have a stroke the disease onset may occur earlier. Does long standing
HTN predispose one to AD? These are questions that people have that are
currently unresolved. Obviously having multiple strokes and AD is not good
and is certainly worse than having one disease. So whether cerebrovascular
disease has a causal relationship to AD or whether cerebrovascular disease
and AD simply share risk factors is unclear right now.
About 25% of healthy elderly have a stroke on MRI or CAT scan for which
they have never had symptoms. Silent strokes occur in a good portion of
the population. We don't know whether that triggers something or is even
important. Currently vascular biology of the brain is extremely important
and not fully worked out in terms of the relationship to degenerative disease.
ARF: What advice can you offer to young investigators in the AD
Mayeux: This is the renaissance of AD research. We know a lot
more, the methods are better, the imaging is helping us sort out who is
going to get AD and who is not. The genetics are very exciting, and AD
is a major public health problem that will only increase in the immediate
future. Unraveling the biology of A-beta, tau and alpha-synuclein will
making it a tremendous time for those involved in applied research. There
will be more tools to ask more and more specific questions. The jump form
bench to the bedside will become easier. So I think that it is quite an
exciting time to work in clinically oriented research especially from the
avenue of biological markers or epidemiology.
Du Y, Dodel RC, Eastwood BJ, Bales KR, Gao F, Lohmuller F, Muller U, Kurz A, Zimmer R, Evans RM, Hake A, Gasser T, Oertel WH, Griffin WS, Paul SM, and Farlow MR. Association of an interleukin 1 alpha polymorphism with Alzheimer's disease [see comments]. Neurology 2000; 55:480-3. Abstract
Grimaldi LM, Casadei VM, Ferri C, Veglia F, Licastro F, Annoni G, Biunno I, De Bellis G, Sorbi S, Mariani C, Canal N, Griffin WS, and Franceschi M. Association of early-onset Alzheimer's disease with an interleukin-1alpha gene polymorphism [see comments]. Ann Neurol 2000; 47:361-5. Abstract
Manly JJ, Merchant CA, Jacobs DM, Small SA, Bell K, Ferin M, and Mayeux R. Endogenous estrogen levels and Alzheimer's disease among postmenopausal women. Neurology 2000;54: 833-7. Abstract
Mayeux R, Ottman R, Maestre G, Ngai C, Tang MX, Ginsberg H, Chun M, Tycko B, and Shelanski M. Synergistic effects of traumatic head injury and apolipoprotein-epsilon 4 in patients with Alzheimer's disease [see comments]. Neurology 1995;45:555-7. Abstract
Nicoll JA, Mrak RE, Graham DI, Stewart J, Wilcock G, MacGowan S, Esiri MM, Murray LS, Dewar D, Love S, Moss T, and Griffin WS. Association of interleukin-1 gene polymorphisms with Alzheimer's disease [see comments]. Ann Neurol 2000;47:365-8. Abstract
Mayeux R, Saunders AM, Shea S, Mirra S, Evans D, Roses AD, Hyman BT, Crain B, Tang M-X, Phelps CH, the Alzheimer's Disease Centers Consortium on APOE and Alzheimer's Disease. Utility of the APOE Genotype in the Diagnosis of Alzheimer's Disease. New Engl J Med 1998; 338; 506-511. Abstract
Stern Y, Gurland B, Tatemichi TK, Tang MX, Wilder D, and Mayeux R. Influence of education and occupation on the incidence of Alzheimer's disease [see comments]. Jama 1994 Apr 6;271, 1004-10. Abstract
Tang MX, Jacobs D, Stern Y, Marder K, Schofield P, Gurland B, Andrews H, and Mayeux R. Effect of oestrogen during menopause on risk and age at onset of Alzheimer's disease [see comments]. Lancet 1996 Aug 17;348, 429-32. Abstract
Richard Mayeux, M.D., is director of the Gertrude H. Sergievsky Center
and co-director of the Taub Institute for Research on Alzheimer's Disease
and the Aging Brain at Columbia University.